Differentiation of n‑1 Positional Isomers in Antisense Oligonucleotides with Orthogonal LCMS Methods

Investigating the origin of critical product-related impurities during solid-phase synthesis is essential to improving the quality of therapeutic oligonucleotides. In the synthesis of a 2′-O-[2-(methylamino)-2-oxoethyl] modified phosphorothioate antisense oligonucleotide (NMA PS ASO), we observed elevated levels of n minus NMA 5-methylcytosine (n-NMA Me C), where n is the full-length product (FLP). This impurity, which results from the deletion of any of the five NMA Me C residues in the oligonucleotide sequence, is the major contributor to the n-1 class of impurities. High-resolution mass spectrometry cannot differentiate positional isomers arising from specific nucleotide deletions, because they are isobaric. Here, we compare two orthogonal approaches for the accurate differentiation of five isomeric n-NMA Me C impurities in a NMA PS ASO sample: (1) desulfurization to enhance chromatographic separation of the individual components of the composite n-NMA Me C impurity, thus enabling straightforward quantitation via LCMS, and (2) direct fragmentation of n-NMA Me C impurities to calculate the content of individual standards using their distinctive MS/MS fragments. Our strategies established the feasibility of resolving up to five isomeric n-1 impurities and allowed us to mitigate the risk of their formation during synthesis. The study results revealed that low coupling efficiency occurred at the NMA Me C nucleotides toward the 5′ end of the sequence, which provided valuable insights into the sequence-specific challenges associated with synthesis efficiency and guided us to implement more effective n-1 control measures during the oligonucleotide synthetic process.